SYMC: svc_monitor scheduling enhancements
Bug #1536352 reported by
Varun Lodaya
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Juniper Openstack |
New
|
High
|
Rudra Rugge | ||
| R2.20 |
New
|
High
|
Rudra Rugge | ||
| R2.21.x |
New
|
High
|
Rudra Rugge | ||
| R2.22.x |
In Progress
|
High
|
Rudra Rugge | ||
Bug Description
svc_monitor process is very slow while scheduling service instance vms. While debugging we saw the following:
1) svc_monitor process makes 1 call to nova to get availability_zone per service_instance vm. This is very unoptimal, specially when the svc_monitor process is coming up.
2) Also, svc_monitor process makes too many API calls for every VM scheduling, need to optimize this.
| information type: | Proprietary → Public |
Revision history for this message
| Nagabhushana R (bhushana) wrote | #1 |
| Changed in juniperopenstack: | |
| importance: | Undecided → High |
| assignee: | nobody → Rudra Rugge (rudrarugge) |
| tags: | added: config service-chain |
| summary: |
- svc_monitor scheduling enhancements + SYMC: svc_monitor scheduling enhancements |
Revision history for this message
| OpenContrail Admin (ci-admin-f) wrote [Review update] R2.22.x | #2 |
To post a comment you must log in.
The 'port-tuple' changes being done in 3.0 addresses these kind of issues. Rudra can add more details.